HIV infection of the genital mucosa in women
The vast majority of new HIV infections are acquired via the genital and rectal mucosa. Here, we provide an overview of our current knowledge of how HIV establishes local infection, with an emphasis on viral invasion through the female genital tract. Studies using human explant tissues and in vivo animal studies have improved our understanding of the cellular and molecular pathways of infection; this information could be harnessed to design effective HIV vaccines and microbicides.
HIV Infection of the
Genital Mucosa in Women
Florian Hladik, MD, PhD, and Thomas J. Hope, PhD
Florian Hladik, MD, PhD
Fred Hutchinson Cancer Research Center, 1100 Fairview
Avenue North, D3-100, Seattle, WA 98109-1024, USA.
Current HIV/AIDS Reports 2009, 6:20–28
Current Medicine Group LLC ISSN 1548-3568
Copyright © 2009 by Current Medicine Group LLC
The vast majority of new HIV infections are acquired
via the genital and rectal mucosa. Here, we provide
an overview of our current knowledge of how HIV
establishes local infection, with an emphasis on
viral invasion through the female genital tract. Stud-
ies using human explant tissues and in vivo animal
studies have improved our understanding of the
cellular and molecular pathways of infection; this
information could be harnessed to design effective
HIV vaccines and microbicides.
The global spread of HIV infection has been fueled by
sexual transmission through the lower genital and rectal
mucosa. These transmission routes account for most cur-
rent and new infections (Table 1). Designing vaccines and
microbicides to prevent HIV-1 acquisition requires a clear
understanding of how HIV establishes initial infection in
the mucosa. Despite the inability to directly observe these
activities in the virus’s natural human host, our knowl-
edge of the molecular and cellular events following viral
exposure has been expanded by combining new virus
detection technologies with improved human ex vivo tis-
sue and in vitro cellular models, as well as with animal
challenge studies in vivo. Our review focuses on the events
immediately following HIV exposure in the lower genital
mucosa of women.
Anatomic Sites for HIV Invasion in the
Female Genital Tract
Although vaginal intercourse carries a lower HIV transmis-
sion probability per exposure event, it contributes more new
HIV cases than anal intercourse or parenteral inoculation
(Table 1). Women appear to be at higher risk for infec-
tion than men, with an estimated 30% to 40% of annual
worldwide infections occurring through HIV invasion of the
female genital tract via exposure to virus-containing semen.
The relative contributions of the vaginal, ectocervical, and
endocervical mucosa to successful transmission remain
unknown, but HIV penetration and infection have been
demonstrated in all three sites. The vaginal mucosa and ecto-
cervix are covered by a multilayered squamous epithelium,
which, if intact, offers better mechanical protection against
pathogen invasion than the single-layer columnar epithelium
of the endocervix. This nding has led to the belief that most
genital HIV transmission in women likely happens through
the endocervix. However, the surface area of the vaginal
wall and ectocervix is approximately 15 times larger than
that of the endocervix  and thus provides more opportuni-
ties for HIV to encounter breaches in the epithelium through
which it more easily penetrates the mucosa. At any rate,
the picture is emerging that in many cases, male-to-female
transmission occurs by only a single particle or a very low
number of viral particles that succeed in crossing the epi-
thelium . The mucosal linings of the female genital tract
usually provide a robust barrier that functions ef ciently to
prevent transmission. Despite this general robustness, it is
conceivable that HIV simultaneously invades more than one
anatomic site. However, retroviral transmission can occur
solely through the vaginal mucosa, as demonstrated by infec-
tions in a woman born without a uterus  and in macaques
after surgical extirpation of the uterus . In fact, exclusive
transmission through the vaginal mucosa may occur quite
regularly, as implied by the nding that women using a dia-
phragm, which insulates the uterine cervix from contact with
HIV-containing semen, were not protected from infection
. However, adherence to correct diaphragm use is dif cult
to validate; thus, some uncertainties remain regarding where
infection occurred in these women.
The question of whether HIV can contact and invade the
endometrium of the uterus by breaching the endocervical
mucus plug has not been well studied. Infection of uterine
cells was detected as early as 2 days after vaginal simian-
HIV inoculation of a single monkey. This was probably too
little time for stromal or lymphatic spread from infection
foci in the lower genital tract to the uterus, thus indicating
that ascent of virus through the endocervical mucus plug
HIV Infection of the Genital Mucosa in Women
Hladik and Hope
may be possible . However, this evidence has not been
corroborated in any more animals or in humans. Thus,
whereas cells in uterine tissue are susceptible to binding
and infection if exposed directly to HIV , transmission
via ascent of the virus through the endocervical canal into
the uterine cavity is not a well-documented event. There-
fore, we focus our discussion on HIV invasion through the
lower female genital tract.
How Does HIV Penetrate the
Genital Mucosa of Women?
HIV infection in vivo can be established by both free and
cell-associated viruses (Fig. 1). Both mechanisms of infection
have been observed in female macaques infected with simian
immunode ciency virus (SIV) , mice infected with HIV
, and indirectly in humans through genetic matching of
HIV viruses sequenced from acutely infected women and
from seminal cells and plasma from their infected partners
. Human cervical explant studies have also con rmed
transmission of cell-free and cell-associated HIV-1 .
Thus, both viral forms can successfully penetrate the mucosa.
Initially, some of the deposited seminal cells or free virions
are trapped in cervical mucus [2,12]. However, it remains
unclear whether this trapping fosters transmission by allow-
ing longer contact time of infected cells or free virions with
the mucosa or whether the trapping impedes transmission by
immobilizing the cells and virions and increasing their ability
to be attacked by innate antiviral substances. Murine studies
revealed that lymphocytes and macrophages inoculated into
the vaginal cavity migrated into the cervicovaginal stroma
and beyond to the iliac lymph nodes . Presumably, HIV
can be carried along with these cells into the draining lym-
phatics. However, transmission from infected donor cells
appears to occur mostly due to localized secretion of virions
upon contact with the genital epithelium [14–16].
HIV virions, either inoculated cell free or released from
infected donor cells, interact with epithelial cells in a variety
of ways. Researchers have used several epithelial cell types
originating from various anatomic locations, as well as
primary and immortalized cell lines. This diversity in cell
types may explain the variability in the ndings. By using
epithelial cells derived from the lower female genital tract,
several groups have demonstrated the binding and entry
of HIV-1, as well as the subsequent ability of HIV-1 to be
transferred to susceptible CD4
T cells [17,18]. Another pro-
cess that may contribute to viral transfer in primary genital
epithelial cells is transcytosis. During this cellular process,
virions in polarized intact epithelial cells can cross from the
apical to the basal region, where they are released and then
can infect susceptible leukocytes [19,20]. However, trans-
cytosis passes on only a very small percentage of the viral
particles that associate with an epithelial cell monolayer
barrier. Interestingly, a virus secreted from infected donor
cells appears more ef cient in transcytosis than a cell-free
virus [15,16,19]. Such newly secreted viruses may even
lead to productive infection of cervical epithelial cells ,
although this nding remains controversial [17,21].
Table 1. Contribution of mucosal HIV invasion sites to global HIV infections in adults
HIV invasion site
to HIV cases
Female genital tract Vagina Squamous,
Semen 1 in 200 to
1 in 2000
Other Various epithelia
Male genital tract Inner foreskin Squamous,
rectal secretions and
1 in 700 to
1 in 3000
Penile urethra Columnar, stratiﬁ ed
Other Various epithelia
Intestinal tract Rectum Columnar,
Semen 1 in 20 to 1 in 300 3.9
Upper GI tract Various epithelia 1 in 2500 1.5
*Includes MSM, bisexual men, and heterosexual men.
Includes MSM, bisexual men, and women infected via anal-receptive intercourse.
GI—gastrointestinal; MSM—men having sex with men.
(Adapted from Hladik and McElrath .)
The Science of HIV Medicine
Figure 1. HIV transmission via the squamous epithelium of the lower female genital tract. Depicted are two stromal papillae. The stromal
papilla on the right of the illustration signiﬁ es the afferent arm shuttling blood cells to the mucosa. Precursor cells of the monocyte lineage
differentiate upon arrival into dendritic cells (DCs) or macrophages. Shearing of the outer epithelium during sexual intercourse lays bare the
tip of the stromal papilla on the left, and possible pathways for HIV invasion within such a “microanatomic infection unit” are indicated by
letters. On the top of the ﬁ gure, HIV receptors and some phenotypic cell receptors are shown. A, Trapping of free HIV virions or HIV-
infected donor cells in mucus covering the mucosa. B, Attachment of HIV-infected donor cells to the luminal surface of the mucosa, which
HIV Infection of the Genital Mucosa in Women
Hladik and Hope
Using an electron microscopy to study isolated vagi-
nal epithelial sheets, we have observed that HIV-1 virions
are sequestered in endocytic organelles and the cytoplasm
of epithelial cells (Hladik, unpublished data). In these
experiments, the epithelial sheets oated freely in virus-
containing media, allowing HIV-1 simultaneous access to
the luminal and basal side of the epithelium. Despite this
dual-sided access, virions were found exclusively in basal
and suprabasal epithelial cells. Thus, more super cial epi-
thelial cells do not appear to be susceptible to endocytosis
and transcytosis, and viral particles are unlikely to be tran-
scytosed through the outer layers of the squamous genital
epithelium. HIV virions either do not enter these cells or
are rapidly degraded before they are passed on to the next
layer of cells. However, HIV-1 has been shown to penetrate
several layers from the luminal surface into the thin gaps
between squamous epithelial cells of the cervicovaginal
mucosa (Hope, unpublished data). This penetration may
bring the virus in direct contact with two cell types that
it may infect later: intraepithelial Langerhans cells (LCs)
T lymphocytes [22•]. In addition, the virus may
reach suprabasal or basal epithelial cells that are susceptible
to viral binding, endocytosis, or transcytosis.
Several cell-surface molecules have been implicated
in HIV-1 binding to genital epithelial cells. Cell-surface
glycosphingolipids—speci cally galactosylceramide on
ectocervical cells  and sulfated lactosylceramide on
vaginal cells —bind HIV-1 gp120. A few studies have
proposed that HIV-1 gp120 interacts with transmembrane
heparan sulfate proteoglycans (syndecans) on genital epi-
thelial cells [18,20]. Another study found that glycoprotein
340—a splice variant of salivary agglutinin expressed on
cervical and vaginal epithelia—speci cally bound the
HIV envelope and enhanced passage of HIV to suscep-
tible leukocytes . One group found that β1-integrin on
explant cervical epithelium bound virions, presumably via
bronectin, which is abundant in human semen and may
coat the envelope of seminal virions, although this nding
was not observed in all explants . Expression analy-
sis of chemokine receptors that bind gp120 on cervical
epithelial cells was not conclusive across several studies.
Thus, chemokine receptors’ involvement in binding of
HIV-1 to epithelial cells remains unclear. Penetration of
virions in between cervicovaginal epithelial cells also can
occur independently of gp120, indicating that penetra-
tion does not necessarily require interaction of the viral
envelope with a speci c receptor (Hope, unpublished
data). Of note, factors in human semen seem to enhance
virus attachment to epithelial cells and leukocytes, thus
increasing infectivity, as has been demonstrated recently
for amyloid brils formed from naturally occurring frag-
ments of seminal prostatic acidic phosphatase [25•].
Whatever mode of viral penetration is used by HIV,
studies with SIV in macaques indicate that virus penetra-
tion into the cervicovaginal epithelium after exposure in
vivo occurs within 30 to 60 minutes . Once in the
epithelium, HIV encounters LCs, a subset of immature
intraepithelial dendritic cells (DCs), and CD4
T cells. The
cellular processes of LCs can reach up to the most super-
cial layers of the epithelial surface , enabling HIV to
directly bind to LCs (Hope, unpublished data). However,
the direct sampling of luminal pathogens by endocervical
DCs or vaginal LCs in vivo, a mechanism that could be
exploited by HIV to bypass the epithelial barrier, has not
been shown directly.
Another penetration mechanism could be that
intercourse-induced mechanical microabrasions of the
mucosal surfaces allow HIV to directly access resident
leukocytes, perhaps even the stromal T cells, DCs, and
macrophages found in the deeper layers of the mucosa.
The vaginal epithelium, like squamous epithelia at other
sites (eg, skin and oral mucosa), is not a structure of uni-
form thickness. Rather, the stromal papillae frequently
project deep into the outer epithelium, creating consider-
able variations in distance between the luminal surface
and the basal membrane. Because of this structure, squa-
mous epithelial barrier thickness can vary from 50 to 250
microns as it protects the underlying stroma. Microabra-
sions in regions of minimally layered epithelium could
more easily expose the basal layers of the epithelium or
even the underlying stroma (Fig. 1). Interestingly, LCs
and, on the stromal side, T cells and macrophages 
cluster at the tips of the stromal projections into the
epithelium, thus creating microareas with an increased
likelihood of infection. The highly focal manner of ini-
tial SIV infection in the genital mucosa of macaques in
Figure 1 (Continued). then secrete virions upon contact. C, Penetration of virions into gaps between epithelial cells. D, Capture of penetrat-
ing virions by Langerhans cells (LCs) residing within the epithelium that extend processes toward the vaginal lumen. E, Internalization of
virions into endocytic compartments of LCs. F, Fusion of HIV with the surface of intraepithelial CD4
T lymphocytes, followed by produc-
tive infection. G, Transcytosis of virions through epithelial cells located close to or within the basal layer of the squamous epithelium. H,
Productive infection of basal epithelial cells. I, Internalization of virions into endocytic compartments of basal epithelial cells. J, Migration
of infected donor cells along physical abrasions of the epithelium into the mucosal stroma. K, Migration of free virions along microabrasions
into the stroma, where they can make direct contact with stromal DCs. L, Productive infection of stromal DCs by HIV. M, Internalization
of virions into endocytic compartments of stromal DC. N, Passage of virus from stromal DCs to CD4
T lymphocytes across an infectious
synapse. O, Massive productive infection of mucosal CD4
T cells activated by contact with antigen-presenting DCs. P, Productive infection
of resting mucosal CD4
memory T cells. Q, Binding of HIV and, possibly, productive infection of stromal macrophages. R, Migration of
productively infected CD4
T cells and DCs into the submucosa and the draining lymphatic and venous microvessels. Migrating DCs may
originate from intraepithelial LCs or stromal DCs. DCs and T cells often form conjugates, and HIV may be passed between the two cell types
along an infectious synapse. DCs carry virions in endocytic compartments; some are also productively infected, but it remains unclear at
which differentiation stage this occurs. DC-SIGN—dendritic cell–speciﬁ c, intercellular adhesion molecule 3–grabbing non-integrin.
The Science of HIV Medicine
vivo  may re ect this notion. In addition to the purely
mechanical microabrasions, chemical and infectious
alterations of the mucosa may open access routes for HIV
to reach resident leukocytes.
Role of Cervicovaginal Langerhans
Cells in HIV Invasion
In studies completed as early as 1987, skin LCs have been
reported to be susceptible to HIV-1 entry and infection
[29,30]. Furthermore, studies have found that LCs in the
genital mucosa of macaques harbor SIV within 24 hours
of intravaginal inoculation ; however, cervicovaginal
LCs were proven only recently to be targets for HIV [22•].
One technical obstacle encountered by researchers using
explant models of whole genital mucosa was that LCs exit
the epithelium shortly after the initiation of organ cultures
[11,21,31,32]. Thus, determining whether the infection of
LCs was occurring within the epithelium was not possible.
One study demonstrated that DCs migrating from HIV-
1–exposed human cervical tissue ef ciently captured and
transmitted the virus in trans, but because whole organ
cultures were used, the authors could not discern whether
the cells originated from the epithelium or the stroma .
We resolved this ambiguity by exposing isolated epithelial
sheets to HIV after the sheets had been separated from the
underlying vaginal stroma. In this way, we demonstrated
that vaginal LCs very ef ciently internalize HIV-1 into
cytoplasmic organelles [22•]. Thus, when the cells subse-
quently exit the epithelium at the basal side, they can take
intact virions with them.
The questions of which receptors are involved in entry
of HIV-1 into vaginal LCs and whether vaginal LCs can
be productively infected remain unanswered. Vaginal LCs
express CD4, CCR5, and the C-type lectin langerin but
not CXCR4 and another C-type lectin named dendritic
cell–speci c, intercellular adhesion molecule 3–grab-
bing non-integrin (DC-SIGN) [22•,33–35]. In our study,
uptake of R5-tropic HIV-1 into vaginal LCs was partially
blocked by antibodies binding CD4 and CCR5, receptors
that normally can mediate viral fusion and productive
infection. However, we could not detect productive infec-
tion of vaginal LCs. In contrast, in LCs from human skin
explants, CD4/CCR5-mediated productive infection has
been observed at low levels . The difference in these
ndings may be explained by the fact that the less sensi-
tive techniques used in our study of vaginal LCs may not
have been able to detect such a low-level infection [22•].
At any rate, if viral production occurs in vaginal LCs, it
seems relatively inef cient compared with HIV’s ability to
be endocytosed by these cells. This does not necessarily
mean that productive HIV-1 infection is irrelevant in vag-
inal LCs; low levels of productive infection in cutaneous
DCs and LCs have been shown to result in extensive viral
replication in cocultured T lymphocytes [30,36]. Future
studies need to determine whether cervicovaginal LCs
support productive infection and whether their infection
can lead to the passage of virus to T cells.
In most DC types, C-type lectins are very good media-
tors of viral entry . However, in a study of vaginal
LCs, the binding and endocytosis of HIV-1 was only
weakly inhibited by mannan, an inhibitor of C-type,
lectin-mediated viral entry [22•]. This nding may not
be that surprising, as HIV-1 is equipped with additional
means of binding to DCs that are independent of C-type
lectins and CCR5 [37,38]. In contrast to vaginal LCs,
HIV is ef ciently captured via the C-type lectin langerin
in epidermal LCs [39•]. Upon viral entry, langerin in epi-
dermal LCs directs HIV-1 to Birbeck granules, where the
virus is degraded. Thus, in vaginal LCs, HIV may bypass
C-type lectins, including langerin, in favor of other routes
of binding and endocytosis, allowing the virus to reach
endocytic compartments that are less hostile to the virus
than the Birbeck granules. In keeping with this notion,
intact virions were still present in LCs derived from the
vaginal epithelium 60 hours after viral challenge [22•].
Further studies are needed to determine which endocytic
pathways are exploited by HIV in vaginal LCs and how
the pathway of entry into the cell may affect the virus’s
ultimate fate in vaginal LCs.
Role of Cervicovaginal CD4
in HIV Invasion
The lamina propria of the human vagina, ectocervix,
and endocervix contains many CD4
T cells, which are
often enriched in the super cial stroma, close to the basal
membrane . CD4
T cells also in ltrate the vaginal
and ectocervical squamous epithelium . Most of these
in ltrating cells are of the CD45RO
and express relatively high levels of CCR5 compared with
T cells found in the peripheral blood [22•,41,42]. Corre-
spondingly, investigators using organ cultures of vaginal,
ectocervical, and endocervical tissues consistently found
high numbers of CD4
T cells that were infected with
HIV-1 beginning 1 day after virus exposure [11,21,32].
Surprisingly, given the presence of CCR5
within the squamous epithelium, HIV-infected T cells
were located exclusively beneath the epithelium in the
mucosal stroma. However, by visualizing the distribu-
tion of virions early (within a few hours after challenge
of the vaginal mucosa), we recently found that R5-tropic
HIV-1 binds and fuses to intraepithelial vaginal CD4
cells very ef ciently, leading to productive infection [22•].
Therefore, infected T cells leave the epithelium rapidly
and consequently are detected in the mucosal stroma if
the analysis is performed later.
Although it is impossible to selectively remove LCs
from the epithelial sheets, several lines of evidence indicate
that initial infection of intraepithelial CD4
T cells is not
HIV Infection of the Genital Mucosa in Women
Hladik and Hope
dependent on LC-mediated viral uptake and infection in
trans [22•]. At any rate, human explant studies concur that
HIV-1 very effectively targets CD4
T cells in the genital
mucosa for productive infection [11,22•,32]. Moreover,
SIV challenge experiments in macaques con rm the impor-
tance of genital CD4
T cells for early retroviral infection
[43,44]. Indeed, after intravenous SIV infection, vaginal
T cells are rapidly depleted in macaques [6,44]. Of
note, SIV productively infects not only activated HLA-DR
T cells but also resting HLA-DR
T cells .
Likewise, our vaginal explant infection model revealed
binding of HIV-1 to HLA-DR
T cells in
the epithelium (Hladik, unpublished data). Calculations
relating relative SIV production to the size of each cell
population in resting versus activated CD4
T cells in the
genital mucosa indicated that infected resting CD4
contribute substantially to viral production .
Role of Other Cervicovaginal Leukocyte
Populations in HIV Invasion
In many of the human explant studies discussed previ-
ously, macrophages in the cervicovaginal stroma were
also identi ed as targets for HIV-1 infection [21,31,46].
In two different organ culture models, macrophages
were the major cell type infected by CCR5-tropic HIV-1
[21,46]. Curiously, however, immunostaining studies have
never conclusively demonstrated that macrophages in
the human uterine cervix or the vagina express CCR5 in
situ. Most macrophages were CCR5 positive when tested
after migration from vaginal organ cultures , but this
CCR5 expression could have been acquired upon leaving
the mucosa . In contrast to experiments in the human
explant model, vaginal SIV challenge experiments in
macaques either rarely revealed SIV-infected macrophages
in genital tissues  or failed to detect them at all [26,49].
These ndings indicate that organ cultures may yield dif-
ferent results than in vivo studies because macrophages
may become activated after harvesting the tissue, leading
to an increase of CCR5 expression and susceptibility to
infection. Therefore, the role of macrophages in early HIV
infection of the lower genital tract in women still needs to
be de ned. Other experimental systems have shown that
macrophages may not only serve as targets for chemokine
receptor–mediated infection but also can capture intact
virions via syndecans . Moreover, viral entry may occur
via a process called macropinocytosis, in which whole
droplets of uid are captured; HIV may be contained in
these uid droplets, allowing the virus to enter the cyto-
plasm without speci c interactions between the viral
envelope and cell-surface receptors . Captured HIV
can be archived within macrophages for several days and
still be transmitted to T cells [50,52]. It will be interesting
to determine whether any of these mechanisms operate in
Stromal DCs are another cell type implicated in
genital HIV invasion. DCs consist of a complex array of
cell subsets that differ in their origin, maturation state,
and anatomic location. For example, stromal DCs in
the human dermis are distinct from epidermal LCs, and
stromal DCs can be further divided into three subpopula-
tions . Many studies have focused on the interaction
between HIV and DCs. Essentially any DC subset that
was speci cally examined as a target for HIV has been
shown to support productive infection—albeit often at
low levels—or passage of virus to other cells in trans, or
both . These ndings have encouraged the view that
DCs in the genital mucosa must be important for sexual
HIV transmission. Although this idea has been con rmed
recently for intraepithelial LCs, the role of stromal DCs
remains enigmatic. In contrast to LCs, stromal DCs are
DC-SIGN positive [34,35], and some studies have found
that they also express CCR5 [47,55]. However, in situ
studies in the human explant model have not identi ed
DCs in the cervicovaginal stroma as foci for productive
HIV infection [11,21,31,46]. In contrast, intravaginal
challenge with SIV in macaques led to detectable infec-
tion of stromal DCs as early as 18 hours postchallenge
[26,49]. Likewise, HIV-infected DCs were present in the
vaginal stroma of asymptomatic HIV-1–infected women
. These ndings suggest that the disparities between
stromal DC infection noted in humans and macaques is
less likely due to inherent differences between the two
species or between SIV and HIV but rather due to differ-
ences in experimental techniques.
In mucosal challenge experiments with macaques, the
animals are inoculated with SIV in vivo, and the infected
tissues are harvested and examined after necropsy without
further in vitro culture. For obvious reasons, this experi-
mental approach is impossible in humans. Thus, to study
HIV, explant human organ cultures are used in which the
HIV challenge is performed in vitro, followed by a culture
period, after which infection is assessed. Some artifacts
may occur in the explant human organ culture system,
one being that DCs migrate into the supernatant before
the culture is terminated and the tissue is xed. This may
be why infected stromal DCs were absent from the whole
organ cultures. Indeed, when DCs were collected from the
supernatant of HIV-1–challenged cervical explant cul-
tures, in trans infectivity could be recovered . Because
the passage of infection from DCs to susceptible target
cells was partially blocked by mannan or anti–DC-SIGN
antibodies, this study determined that C-type lectins were
involved in the uptake of virus by the migratory DCs.
However, this study did not examine productive infection
of the migratory cells. In our own experiments, we have
investigated productive infection of human cervicovaginal
DCs by rst harvesting the cells that are migrating from
cervicovaginal organ cultures and then exposing them to
HIV-1 . Our results showed that massive budding of
The Science of HIV Medicine
virions occurred in the migrated DCs 5 days after virus
exposure . However, in both studies, the origin of the
DCs from either the epithelium proper or the submucosal
stromal tissue remained unclear. Moreover, by challenging
cells with HIV after migration, as performed in the second
study, the cells’ phenotype may have changed during the
migration period, rendering the experimental conditions
less representative of in vivo infection. Regardless, this
experiment unequivocally proved that under certain (but
not yet well-de ned) conditions, genital DCs can become
highly ef cient producers of viral progeny. Clearly, much
remains to be learned about the general biology of stro-
mal DCs in the human genital mucosa; for example, we
need to determine whether similar DC subsets exist as in
the dermis  and de ne the nature of the interaction
between these subsets and HIV.
HIV has been shown to interact with additional leuko-
cyte populations. Infection of natural killer cells has been
reported , as has been the passage of virus from DC-
SIGN–expressing B lymphocytes to T cells . Whether
natural killer and B cells are important for HIV invasion
in the genital mucosa remains to be investigated. Finally, a
study demonstrating that monocytic precursor cells enter
the mouse dermis in massive numbers upon intracutaneous
leishmania inoculation and differentiate into stromal DCs
 raises the possibility that such an emergency mecha-
nism in response to an infectious insult may also operate
in the human mucosa. Upon their in ux into the genital
mucosa and during differentiation, these monocytic pre-
cursors could present another target for HIV.
Although animal and human tissue explant models have
already provided invaluable insights into the pathways
of mucosal HIV infection, many questions remain.
Whereas the strong protective effect of circumcision pin-
points the foreskin as the major site for HIV invasion in
men, in women, the relative contributions to infection
of the vagina, ectocervix, and endocervix are unknown.
On a cellular level, the role of LCs and stromal DCs in
the spread of HIV-1 infection in the mucosa remains ill
de ned. In vitro studies have demonstrated that DCs can
effectively pass HIV-1 across an infectious synapse to
T cells , but this concept is still unproven in the
genital mucosa. Likewise, whether endocytosis of HIV
into mucosal LCs or DCs leads to preservation of viable
virions or to viral degradation is a question with important
prophylactic implications. If endocytosed virions enter a
dead-end pathway in LCs, de ning the exact pathways of
endocytosis—in order to design effective blocking strate-
gies—becomes a less urgent requirement for developing a
prevention strategy. Nevertheless, HIV’s ability to infect
multiple types of leukocytes along its path from the muco-
sal surface to the draining lymphatics is likely a de ning
feature of its biologic transmission tness. The dif culty
of preventing HIV transmission with a vaccine or micro-
bicide may be in part attributable to the parallel pathways
that HIV may travel to establish mucosal infection.
The authors wish to thank Reneé Ireton for her editorial
No potential con icts of interest relevant to this article
References and Recommended Reading
Papers of particular interest, published recently,
have been highlighted as:
• Of importance
•• Of major importance
1. Pendergrass PB, Belovicz MW, Reeves CA: Surface area
of the human vagina as measured from vinyl polysiloxane
casts. Gynecol Obstet Invest 2003, 55:110–113.
2. Miller CJ, Li Q, Abel K, et al.: Propagation and dissemi-
nation of infection after vaginal transmission of simian
immunode ciency virus. J Virol 2005, 79:9217–9227.
3. Kell PD, Barton SE, Edmonds DK, Boag FC: HIV infec-
tion in a patient with Meyer-Rokitansky-Kuster-Hauser
syndrome. J R Soc Med 1992, 85:706–707.
4. Miller CJ, Alexander NJ, Vogel P, et al.: Mechanism of
genital transmission of SIV: a hypothesis based on transmis-
sion studies and the location of SIV in the genital tract
of chronically infected female rhesus macaques. J Med
Primatol 1992, 21:64–68.
5. Padian NS, van der Straten A, Ramjee G, et al.: Diaphragm
and lubricant gel for prevention of HIV acquisition in
southern African women: a randomised controlled trial.
Lancet 2007, 370:251–261.
6. Joag SV, Adany I, Li Z, et al.: Animal model of muco-
sally transmitted human immunode ciency virus type 1
disease: intravaginal and oral deposition of simian/human
immunode ciency virus in macaques results in systemic
infection, elimination of CD4+ T cells, and AIDS. J Virol
7. Howell AL, Edkins RD, Rier SE, et al.: Human immunode-
ciency virus type 1 infection of cells and tissues from the
upper and lower human female reproductive tract. J Virol
8. Kaizu M, Weiler AM, Weisgrau KL, et al.: Repeated
intravaginal inoculation with cell-associated simian
immunode ciency virus results in persistent infection of
nonhuman primates. J Infect Dis 2006, 194:912–916.
9. Khanna KV, Whaley KJ, Zeitlin L, et al.: Vaginal transmis-
sion of cell-associated HIV-1 in the mouse is blocked by a
topical, membrane-modifying agent. J Clin Invest 2002,
10. Zhu T, Wang N, Carr A, et al.: Genetic characterization
of human immunode ciency virus type 1 in blood and
genital secretions: evidence for viral compartmentalization
and selection during sexual transmission. J Virol 1996,
HIV Infection of the Genital Mucosa in Women
Hladik and Hope
11. Gupta P, Collins KB, Ratner D, et al.: Memory CD4(+) T
cells are the earliest detectable human immunode ciency
virus type 1 (HIV-1)-infected cells in the female genital
mucosal tissue during HIV-1 transmission in an organ
culture system. J Virol 2002, 76:9868–9876.
12. Maher D, Wu X, Schacker T, et al.: HIV binding, penetra-
tion, and primary infection in human cervicovaginal tissue.
Proc Natl Acad Sci U S A 2005, 102:11504–11509.
13. Ibata B, Parr EL, King NJ, Parr MB: Migration of foreign
lymphocytes from the mouse vagina into the cervicovaginal
mucosa and to the iliac lymph nodes. Biol Reprod 1997,
14. Phillips DM, Tan X, Perotti ME, Zacharopoulos VR: Mecha-
nism of monocyte-macrophage-mediated transmission of HIV.
AIDS Res Hum Retroviruses 1998, 14(Suppl 1):S67–S70.
15. Alfsen A, Yu H, Magerus-Chatinet A, et al.: HIV-1-infected
blood mononuclear cells form an integrin- and agrin-depen-
dent viral synapse to induce ef cient HIV-1 transcytosis across
epithelial cell monolayer. Mol Biol Cell 2005, 16:4267–4279.
16. Van Herrewege Y, Michiels J, Waeytens A, et al.: A dual
chamber model of female cervical mucosa for the study of
HIV transmission and for the evaluation of candidate HIV
microbicides. Antiviral Res 2007, 74:111–124.
17. Dezzutti CS, Guenthner PC, Cummins JE Jr, et al.: Cervical
and prostate primary epithelial cells are not productively
infected but sequester human immunode ciency virus type
1. J Infect Dis 2001, 183:1204–1213.
18. Wu Z, Chen Z, Phillips DM: Human genital epithelial
cells capture cell-free human immunode ciency virus type
1 and transmit the virus to CD4+ cells: implications for
mechanisms of sexual transmission. J Infect Dis 2003,
19. Bomsel M: Transcytosis of infectious human immunode -
ciency virus across a tight human epithelial cell line barrier.
Nat Med 1997, 3:42–47.
20. Bobardt MD, Chatterji U, Selvarajah S, et al.: Cell-free
human immunode ciency virus type 1 transcytosis through
primary genital epithelial cells. J Virol 2007, 81:395–405.
21. Greenhead P, Hayes P, Watts PS, et al.: Parameters of
human immunode ciency virus infection of human cervical
tissue and inhibition by vaginal virucides. J Virol 2000,
22.• Hladik F, Sakchalathorn P, Ballweber L, et al.: Initial
events in establishing vaginal entry and infection by
human immunode ciency virus type-1. Immunity 2007,
This is the rst analysis of the infection events taking place in the
outer epithelium of the human vagina shortly after exposure to
HIV-1. This analysis found that intraepithelial CD4
T cells are
productively infected by viral fusion, whereas LCs endocytose
virions, a mechanism that did not lead to measurable productive
infection in the LCs.
23. Furuta Y, Eriksson K, Svennerholm B, et al.: Infection of
vaginal and colonic epithelial cells by the human immuno-
de ciency virus type 1 is neutralized by antibodies raised
against conserved epitopes in the envelope glycoprotein
gp120. Proc Natl Acad Sci U S A 1994, 91:12559–12563.
24. Stoddard E, Cannon G, Ni H, et al.: gp340 expressed on
human genital epithelia binds HIV-1 envelope protein
and facilitates viral transmission. J Immunol 2007,
25.• Munch J, Standker L, Adermann K, et al.: Discovery and
optimization of a natural HIV-1 entry inhibitor targeting
the gp41 fusion peptide. Cell 2007, 129:263–275.
Amyloidogenic fragments of prostatic acidic phosphatase in semen
serve as enhancing factors for HIV infection.
26. Hu J, Gardner MB, Miller CJ: Simian immunode ciency
virus rapidly penetrates the cervicovaginal mucosa after
intravaginal inoculation and infects intraepithelial dendritic
cells. J Virol 2000, 74:6087–6095.
27. Miller CJ, McChesney M, Moore PF: Langerhans cells,
macrophages and lymphocyte subsets in the cervix and
vagina of rhesus macaques. Lab Invest 1992, 67:628–634.
28. Yeaman GR, Asin S, Weldon S, et al.: Chemokine recep-
tor expression in the human ectocervix: implications for
infection by the human immunode ciency virus-type I.
Immunology 2004, 113:524–533.
29. Tschachler E, Groh V, Popovic M, et al.: Epidermal Lang-
erhans cells—a target for HTLV-III/LAV infection. J Invest
Dermatol 1987, 88:233–237.
30. Kawamura T, Cohen SS, Borris DL, et al.: Candidate
microbicides block HIV-1 infection of human immature
Langerhans cells within epithelial tissue explants. J Exp Med
31. Collins KB, Patterson BK, Naus GJ, et al.: Development
of an in vitro organ culture model to study transmission
of HIV-1 in the female genital tract. Nat Med 2000,
32. Hu Q, Frank I, Williams V, et al.: Blockade of attachment
and fusion receptors inhibits HIV-1 infection of human
cervical tissue. J Exp Med 2004, 199:1065–1075.
33. Turville SG, Cameron PU, Handley A, et al.: Diversity
of receptors binding HIV on dendritic cell subsets. Nat
Immunol 2002, 3:975–983.
34. Geijtenbeek TB, Kwon DS, Torensma R, et al.: DC-SIGN, a
dendritic cell-speci c HIV-1-binding protein that enhances
trans-infection of T cells. Cell 2000, 100:587–597.
35. Jameson B, Baribaud F, Pohlmann S, et al.: Expression of
DC-SIGN by dendritic cells of intestinal and genital mucosae
in humans and rhesus macaques. J Virol 2002, 76:
36. Pope M, Gezelter S, Gallo N, et al.: Low levels of HIV-1
infection in cutaneous dendritic cells promote extensive
viral replication upon binding to memory CD4+ T cells.
J Exp Med 1995, 182:2045–2056.
37. Gummuluru S, Rogel M, Stamatatos L, Emerman M: Bind-
ing of human immunode ciency virus type 1 to immature
dendritic cells can occur independently of DC-SIGN and
mannose binding C-type lectin receptors via a cholesterol-
dependent pathway. J Virol 2003, 77:12865–12874.
38. Boggiano C, Manel N, Littman DR: Dendritic cell-medi-
ated trans-enhancement of human immunode ciency virus
type 1 infectivity is independent of DC-SIGN. J Virol 2007,
39.• de Witte L, Nabatov A, Pion M, et al.: Langerin is a natural
barrier to HIV-1 transmission by Langerhans cells. Nat Med
Skin LCs are a dead-end pathway for HIV because the C-type lec-
tin langerin on these cells binds and internalizes HIV into Birbeck
granules, where the virions are degraded.
40. Johansson EL, Rudin A, Wassen L, Holmgren J: Distribu-
tion of lymphocytes and adhesion molecules in human
cervix and vagina. Immunology 1999, 96:272–277.
41. Hladik F, Lentz G, Delpit E, et al.: Coexpression of CCR5
and IL-2 in human genital but not blood T cells: implications
for the ontogeny of the CCR5+ Th1 phenotype. J Immunol
42. Zhang L, He T, Talal A, et al.: In vivo distribution of the
human immunode ciency virus/simian immunode ciency
virus coreceptors: CXCR4, CCR3, and CCR5. J Virol 1998,
43. Zhang Z, Schuler T, Zupancic M, et al.: Sexual transmis-
sion and propagation of SIV and HIV in resting and
activated CD4+ T cells. Science 1999, 286:1353–1357.
44. Veazey RS, Marx PA, Lackner AA: Vaginal CD4+ T cells
express high levels of CCR5 and are rapidly depleted in
simian immunode ciency virus infection. J Infect Dis 2003,
45. Zhang ZQ, Wietgrefe SW, Li Q, et al.: Roles of substrate
availability and infection of resting and activated CD4+ T
cells in transmission and acute simian immunode ciency virus
infection. Proc Natl Acad Sci U S A 2004, 101:5640–5645.
46. Cummins JE Jr, Guarner J, Flowers L, et al.: Preclinical
testing of candidate topical microbicides for anti-human
immunode ciency virus type 1 activity and tissue toxicity
in a human cervical explant culture. Antimicrob Agents
Chemother 2007, 51:1770–1779.
The Science of HIV Medicine
47. Hladik F, Lentz G, Akridge RE, et al.: Dendritic cell-T-
cell interactions support coreceptor-independent human
immunode ciency virus type 1 transmission in the human
genital tract. J Virol 1999, 73:5833–5842.
48. Tuttle DL, Harrison JK, Anders C, et al.: Expression of
CCR5 increases during monocyte differentiation and directly
mediates macrophage susceptibility to infection by human
immunode ciency virus type 1. J Virol 1998, 72:4962–4969.
49. Spira AI, Marx PA, Patterson BK, et al.: Cellular targets of
infection and route of viral dissemination after an intra-
vaginal inoculation of simian immunode ciency virus into
rhesus macaques. J Exp Med 1996, 183:215–225.
50. Bobardt MD, Saphire AC, Hung HC, et al.: Syndecan
captures, protects, and transmits HIV to T lymphocytes.
Immunity 2003, 18:27–39.
51. Marechal V, Prevost MC, Petit C, et al.: Human immuno-
de ciency virus type 1 entry into macrophages mediated by
macropinocytosis. J Virol 2001, 75:11166–11177.
52. Sharova N, Swingler C, Sharkey M, Stevenson M: Macro-
phages archive HIV-1 virions for dissemination in trans.
Embo J 2005, 24:2481–2489.
53. Nestle FO, Nickoloff BJ: Deepening our understand-
ing of immune sentinels in the skin. J Clin Invest 2007,
54. Wu L, KewalRamani VN: Dendritic-cell interactions with
HIV: infection and viral dissemination. Nat Rev Immunol
55. Prakash M, Kapembwa MS, Gotch F, Patterson S: Che-
mokine receptor expression on mucosal dendritic cells
from the endocervix of healthy women. J Infect Dis 2004,
56. Bhoopat L, Eiangleng L, Rugpao S, et al.: In vivo identi ca-
tion of Langerhans and related dendritic cells infected
with HIV-1 subtype E in vaginal mucosa of asymptomatic
patients. Mod Pathol 2001, 14:1263–1269.
57. Harada H, Goto Y, Ohno T, et al.: Proliferative activation
up-regulates expression of CD4 and HIV-1 co-receptors
on NK cells and induces their infection with HIV-1. Eur J
Immunol 2007, 37:2148–2155.
58. Rappocciolo G, Piazza P, Fuller CL, et al.: DC-SIGN on
B lymphocytes is required for transmission of HIV-1 to T
lymphocytes. PLoS Pathog 2006, 2:e70.
59. Leon B, Lopez-Bravo M, Ardavin C: Monocyte-derived
dendritic cells formed at the infection site control the induc-
tion of protective T helper 1 responses against Leishmania.
Immunity 2007, 26:519–531.
60. McDonald D, Wu L, Bohks SM, et al.: Recruitment of HIV
and its receptors to dendritic cell-T cell junctions.
61. Hladik F, McElrath MJ: Setting the stage: host invasion by
HIV. Nat Rev Immunol 2008, 8:447–457.